Merge pull request #5588 from aajanki/encode_frag_filenames

[youtube-dl] / youtube_dl / extractor / youtube.py

diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py
index 887c46d95dcef1febe45166e9dbd2a4c49083e73..030ec70ca0b89c0c910e051c84e26d8d408f458f 100644 (file)
--- a/youtube_dl/extractor/youtube.py
+++ b/youtube_dl/extractor/youtube.py
@@ -202,7 +202,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                                  v=
                              )
                          ))
-                         |youtu\.be/                                          # just youtu.be/xxxx
+                         |(?:
+                            youtu\.be|                                        # just youtu.be/xxxx
+                            vid\.plus                                         # or vid.plus/xxxx
+                         )/
                          |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
                          )
                      )?                                                       # all until now is optional -> you can pass the naked ID
@@ -624,6 +627,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             'params': {
                 'skip_download': True,
             },
+        },
+        {
+            'url': 'http://vid.plus/FlRa-iH7PGw',
+            'only_matching': True,
         }
     ]
 
@@ -653,7 +660,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
 
     def _extract_signature_function(self, video_id, player_url, example_sig):
         id_m = re.match(
-            r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
+            r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?)?\.(?P<ext>[a-z]+)$',
             player_url)
         if not id_m:
             raise ExtractorError('Cannot identify player %r' % player_url)
@@ -1236,7 +1243,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
             if 'rtmpe%3Dyes' in encoded_url_map:
                 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
-            url_map = {}
+            formats = []
             for url_data_str in encoded_url_map.split(','):
                 url_data = compat_parse_qs(url_data_str)
                 if 'itag' not in url_data or 'url' not in url_data:
@@ -1282,7 +1289,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                                 player_desc = 'flash player %s' % player_version
                             else:
                                 player_version = self._search_regex(
-                                    r'html5player-([^/]+?)(?:/html5player)?\.js',
+                                    r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
                                     player_url,
                                     'html5 player', fatal=False)
                                 player_desc = 'html5 player %s' % player_version
@@ -1296,8 +1303,50 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     url += '&signature=' + signature
                 if 'ratebypass' not in url:
                     url += '&ratebypass=yes'
-                url_map[format_id] = url
-            formats = _map_to_format_list(url_map)
+
+                # Some itags are not included in DASH manifest thus corresponding formats will
+                # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
+                # Trying to extract metadata from url_encoded_fmt_stream_map entry.
+                mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
+                width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
+                dct = {
+                    'format_id': format_id,
+                    'url': url,
+                    'player_url': player_url,
+                    'filesize': int_or_none(url_data.get('clen', [None])[0]),
+                    'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
+                    'width': width,
+                    'height': height,
+                    'fps': int_or_none(url_data.get('fps', [None])[0]),
+                    'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
+                }
+                type_ = url_data.get('type', [None])[0]
+                if type_:
+                    type_split = type_.split(';')
+                    kind_ext = type_split[0].split('/')
+                    if len(kind_ext) == 2:
+                        kind, ext = kind_ext
+                        dct['ext'] = ext
+                        if kind in ('audio', 'video'):
+                            codecs = None
+                            for mobj in re.finditer(
+                                    r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
+                                if mobj.group('key') == 'codecs':
+                                    codecs = mobj.group('val')
+                                    break
+                            if codecs:
+                                codecs = codecs.split(',')
+                                if len(codecs) == 2:
+                                    acodec, vcodec = codecs[0], codecs[1]
+                                else:
+                                    acodec, vcodec = (codecs[0], 'none') if kind == 'audio' else ('none', codecs[0])
+                                dct.update({
+                                    'acodec': acodec,
+                                    'vcodec': vcodec,
+                                })
+                if format_id in self._formats:
+                    dct.update(self._formats[format_id])
+                formats.append(dct)
         elif video_info.get('hlsvp'):
             manifest_url = video_info['hlsvp'][0]
             url_map = self._extract_from_m3u8(manifest_url, video_id)
@@ -1762,7 +1811,7 @@ class YoutubeSearchURLIE(InfoExtractor):
             r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML')
 
         part_codes = re.findall(
-            r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code)
+            r'(?s)<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*>(.*?)</h3>', result_code)
         entries = []
         for part_code in part_codes:
             part_title = self._html_search_regex(